Modular Automation Design for Equipment Management

Fabiano Santos, Bruno Prado, Daniel Dantas and Kalil Bispo

Department of Computing, Federal University of Sergipe, Marechal Rondon Avenue, S

ao Crist

ao, Brazil

Keywords:

Residential Automation, Internet of Things, Wireless Sensor Network, Modular Design.

Abstract:

This paper presents a modular and low cost residential automation solution, using the Internet of Things

(IoT) concept, for control and monitoring electronic equipment. The development of this work is divided in

two parts: the conﬁguration of electronic devices and use of Home Assistant platform. The Home Assistant

platform allows the control and monitoring of electronic equipment over the Internet. It uses the MQTT

communication protocol for integration with the devices. An auxiliary conﬁguration application has been

developed to conﬁgure dynamically the pins of the sensors and actuators. An electronic device was developed

which is responsible for reading and controlling a set of sensors and actuators. The results of the proposed

project were positive. It supports a larger number of sensors and has the lowest cost, of US$1.78.

1 INTRODUCTION

Through personal computers, the Internet, mobile te-

lephony and other technologies that have emerged in

the world of personal computing, the acceptance of

residential technologies has come to have a strong ap-

peal (Muratori and B, 2017). With the creation of the

Internet and in conjunction with residential automa-

tion, it was possible to connect the automated devices

used daily to the Internet. This set of Internet ena-

bled devices is what we call the Internet of Things

(IoT) (Braga, 2017a).

The Internet of Things is considered a technolo-

gical revolution and more and more new products, of

the most varied types such as computers and smartp-

hones, that can be connected to the Internet and other

devices, are to be found (Zambarda, 2014). However,

its application in people’s daily lives, or even access

to it, seems to belong more to the future than to the

present (Freitas, 2016). There are some factors that

are responsible for this, such as: incompatibility of

technologies for the operation of a single system, high

cost of investment, mobile Internet infra-structure be-

low expected and complexity of use and installation

of systems.

There are several home automation solutions, but

many are not standardized. When a solution provides

some sort of standard, it has a high cost and is not

compatible with other market solutions. Another im-

portant factor is the complexity of using and maintai-

ning the systems. It is necessary to create standards

that facilitate the connection between devices from

different manufacturers. Users generally want to con-

trol their devices from a single platform and not have

one for each device, as it currently occurs (Takahashi,

2016).

This article aims to develop a modular and low-

cost residential automation system using the Internet

of Things (Kelly et al., 2013) concept for the control

of electronic equipment, such as light bulbs and hou-

sehold appliances, through the Internet.

2 RELATED WORKS

The work developed by (Mikhaylov and Huttunen,

2014) (Work 1) presents a system of modular elec-

tronic devices, using the concept of Wireless Sen-

sor and Actor Network (WSAN), with connectivity

between plug & play modules. Because it is scala-

ble, it enables the customization of the modules for

the desired application, and together with the plug

& play connection, makes it easy to use. The pro-

posal of this approach allows the system to be sca-

lable through an IMP-bus bus, which made the har-

dware relatively large and with custom components

that make design expensive. The purpose of the work

of (Mikhaylov and Paatelma, 2015) (Work 2) was

to develop the software architecture for the modular

hardware platform introduced in the (Mikhaylov and

Huttunen, 2014) project. With the proposed architec-

ture, it was possible to support the dynamism of the

Santos, F., Prado, B., Dantas, D. and Bispo, K.

Modular Automation Design for Equipment Management.

DOI: 10.5220/0006701506010606

In Proceedings of the 20th International Conference on Enterprise Information Systems (ICEIS 2018), pages 601-606

ISBN: 978-989-758-298-1

601

hardware platform. For each available peripheral in

the module, a driver is required, which increases the

computational load for the devices. With the use of

a FreeRTOS operating system, the need for a micro-

controller with greater processing power and memory

has increased, since only the RTOS kernel requires

about 5 to 10 KB of space in ROM (FreeRTOS, 2017)

memory, cost of the project.

In the work of (Chi et al., 2014) (Work 3) a de-

vice was designed with a reconﬁgurable interface of

intelligent sensors for industrial Wireless Sensor Net-

works (WSN) in the IoT environment. The system

uses the IEEE 1451 (Song and Lee, 2008) interface

standard, which allows the system to collect sensor

data intelligently. The interface is plug & play, com-

patible with various types of sensors. As this system

is geared towards the industrial environment, it requi-

res more robust components, making the system rela-

tively large, and the use of the ZigBee standard, ma-

kes the project more expensive.

The work of (Suh and Ko, 2008) (Work 4) ad-

dresses an intelligent home control system, based

on WSN. The system has hardware modularity and

through several additional modules built, allows cus-

tomization of the platform for the desired application.

Through a single module, we could allow the con-

nection of a set of predeﬁned sensors and there would

be no need for additional modules, which would sim-

plify the maintenance of the hardware if any sensors

are defective or need to be updated.

The (Kruger et al., 2013) project (Work 5) pro-

poses a new platform related to WSN. The platform

uses a modular design approach, supporting two ope-

rating systems: TinyOS and ContikiOS. In addition, it

is composed of two modular units, one of processing

and communication, another of storage and energy

harvesting. The inclusion of the solar energy harves-

ting module facilitated the use of the platform in harsh

environments where there is no electricity. The sen-

sors were not mentioned and their modules were not

designed, which limits the use of the platform.

The project of (Kelly et al., 2013) (Work 6) pre-

sents a residential automation system for the moni-

toring of environmental conditions, energy manage-

ment and control of household devices, such as lamps

and water heater, using the IoT concept. The har-

dware for data acquisition was built in a modular way,

where each type of detection unit has a speciﬁc pur-

pose. In addition, the system controls the applian-

ces automatically, based on some parameters such as:

sensors, use of the inhabitants and the value of the

energy tariff. Using the ZigBee standard, it was pos-

sible to create a wireless sensor network and make the

system scalable, but made the project more expensive.

Supported modules and sensors are limited, allowing

no further customization of the sensors.

The purpose of (Higuera et al., 2015) work

(Work 7) is to build an intelligent interoperable

lighting solution, combining different technologies.

Through human-focused lighting studies, the system

can change the intensity of light to increase visual

comfort. In addition, this system follows the guideli-

nes deﬁned by ISO/IEC/IEEE 21451 (formerly IEEE

1451) and ZigBee Light Link (Wang, 2013) stan-

dards. The use of the IEEE 21451 standard, which re-

quired additional components, and the ZigBee, made

the project more expensive.

In the work of (Magno et al., 2015) (Work 8) a

new system was proposed to control LED lighting

with a wireless sensor network. The use of light and

motion sensors in combination with user preferences

allowed distributed intelligence to save energy by re-

ducing the intensity of light. Using the ZigBee stan-

dard, the system has become ﬂexible and scalable, but

has become more expensive. In addition, the system

only supports two types of sensors, which limits its

use.

3 SYSTEM DESCRIPTION

The development of this work is divided into two

parts: the Home Assistant platform conﬁguration and

usage and the proposed electronic device develop-

ment.

Figure 1 shows the project architecture, which

consists of the server (Mosquitto broker + Home As-

sistant) and the electronic device (Arduino Nano +

ESP8266). The data communication between server

and electronic device is done over WiFi, using the

MQTT protocol. The server can control actuators and

monitor sensors.

3.1 Home Assistant Platform

Home Assistant is an open source home automation

platform developed in Python 3 (Assistant, 2017b).

This platform provides control, monitoring and auto-

mation of devices, in addition to being able to run

on the most used operating systems, such as Win-

dows, Linux and MacOS, it supports more than 750

components, including MQTT, ZigBee and Z-Wave.

Home Assistant has, as frontend, a web application

with a single interface for mobile devices where you

can control all the equipment. When you start the

application for the ﬁrst time, a default conﬁguration

ﬁle named conﬁguration.yaml is created. This is the

main application ﬁle, which enables the web interface

ICEIS 2018 - 20th International Conference on Enterprise Information Systems

602

Figure 1: Project architecture.

and device discovery. All components (sensors, actu-

ators, etc.) and their settings, basic settings and com-

munication are contained in this ﬁle.

3.1.1 Communication using the MQTT Protocol

For communication between the devices and the

Home Assistant, the MQTT protocol version 3.1

with the broker Mosquitto (Eclipse, 2017) was used.

MQTT uses the publish/subscriber paradigm, where

an element of the network subscribes when it wants

to receive data and publishes when wants to send

data(Barros, 2015). Message identiﬁcation is done

through topics, with levels separated by bars. For

this project, topics were created for each sensor and

actuator, where data sent by the devices is stored.

The default name convention for creating these to-

pics is [IP]/assistant.residential/[sensor or actuator].

The IP corresponds to that of the device where

the sensor or actuator is connected. This makes

it easy to control the components that will be as-

sociated with the device. The actuator is control-

led by the ON and OFF commands, sent to the

[IP]/assistant.residential/control topic. For the sen-

sors, four types of topics were created:

• [IP]/assistant.residential/luminosity: stores the

brightness value provided by the LDR sensor;

• [IP]/assistant.residential/temperature: stores

the temperature provided by the DHT11 sensor;

• [IP]/assistant.residential/humidity: stores the

humidity of the air supplied by the DHT11 sen-

sor;

• [IP]/assistant.residential/movement: stores a

ﬂag that shows if there is movement, according

to the PIR sensor DYP-ME003.

The Home Assistant accesses these information

through its conﬁguration ﬁle, where all the topics are,

each related to its component, as discussed in the

3.1.2.

3.1.2 Device Conﬁguration

Home Assistant has a platform conﬁguration ﬁle,

conﬁguration.yaml, where all components are conﬁ-

gured. For each new component that will be used, you

need to edit this ﬁle with your settings for the platform

to recognize it. In order to facilitate user conﬁguration

and use, a web application was developed in Python

with CGI (Common Gateway Interface) (Foundation,

2017) that will automatically generate this ﬁle. In ad-

dition, this application will allow the customization

of the devices and their components, conﬁguring the

pins of the sensors and actuators. Several devices can

be added, which will be displayed in a table in the

main screen of that application, with the option to edit

them.

3.2 Proposed Electronic Device

The developed electronic device is responsible for re-

ading and controlling a set of certain sensors and ac-

tuators. For this device were chosen components that

met the objectives of the project: low cost and easy

availability, such as the Arduino Nano and the module

ESP8266. These components, plus the set of sensors

and actuators supported, form the electronic device of

this project.

3.2.1 Hardware Development

The Arduino Nano is a free hardware electronic pro-

totyping platform that uses an ATmega328 microcon-

troller (Braga, 2017b). It was chosen for having a

cost of US$1.78 (AliExpress, 2017f) and providing

support for a wide range of sensors, being responsi-

ble for reading these sensors and driving the devices

through the actuators. The ESP8266 module was used

because it provides the Arduino Nano wireless com-

munication (Thomsen, 2015). This module commu-

nicates with the Arduino via serial interface. With it,

the Arduino can connect in 802.11 b/g/n wireless net-

works, operating in AP (Access Point) or STA (Sta-

tion) modes. There are several models, but the one

Modular Automation Design for Equipment Management

603

chosen was the ESP-01 because it is compact, 25 mm

high and 15 mm wide (Thomsen, 2017), and has a

cost of US$2.21 (AliExpress, 2017d). Three types

of sensors and one actuator were used, all of which

are easily available and supported by Arduino Nano.

The three types of sensors used are: brightness sensor

LDR (FilipeFlop, 2017c), temperature and humidity

sensor DHT11 (e Cia, 2013) and the PIR motion sen-

sor (e Cia, 2014). The relay module of a channel was

used as actuator (Thomsen, 2013).

3.2.2 Firmware Development

The ﬁrmware of the electronic device, that is, the Ar-

duino Nano and the ESP8266, was developed. The

Arduino Nano has the function of reading the sensor

data and actuating the actuators, as well as communi-

cating with the ESP8266 module. This communica-

tion was made through the serial interface. In the dia-

gram of Figure 2 you can observe the execution ﬂow

of your ﬁrmware. The Arduino sends to the ESP8266

sensor data and actuator status, previously conﬁgu-

red when adding the device to the application. In this

communication, we observed that the exchange of se-

veral messages in a short time caused losses of cha-

racters, thus, we created a message pattern. This pat-

tern consists of several concatenated messages, that

is, the values of each sensor and actuator are conca-

tenated and sent at one time. A function has been

implemented that associates the sensor/actuator to the

pin conﬁgured in the developed application and stores

the pin value in the EEPROM memory of the Arduino

Nano. This pin is stored in the memory address of the

sensor type. The ESP8266 receives from the conﬁgu-

ration application the pins of the sensors and the ac-

tuator used in the device and sends it to the Arduino

Nano. After the pins are conﬁgured, Arduino starts

reading the connected components and sends its data

to the ESP8266.

The ESP8266 module is responsible for providing

wireless communication to the Arduino Nano. In the

diagram of Figure 3 is shown the execution ﬂow of

its ﬁrmware. It connects to the Wi-Fi network and

also to the Mosquitto broker, establishing communi-

cation between the device and the Home Assistant.

The WiFiManager library, which implements all the

conﬁguration features of the (Tzapu, 2017) network,

was used. These settings are made through a web ap-

plication, included in the library itself. In addition to

the Wi-Fi network conﬁguration, the library provides

conﬁguration of the connection to the Mosquitto bro-

ker.

Figure 2: Arduino Nano ﬁrmware execution ﬂow.

Figure 3: ESP8266 ﬁrmware execution ﬂow.

4 RESULTS AND DISCUSSIONS

All the analyzed works are scalable, allowing the ex-

pansion to control more equipment. In the Table 1,

one can observe a comparative analysis of all the

works and their quantitative parameters. For the plat-

form was estimated the cost of the component used

as the main processor of the device. Cost estimates

were made through prices researched at virtual stores,

giving preference to the website AliExpress (AliEx-

press, 2017c). Some components were not found on

ICEIS 2018 - 20th International Conference on Enterprise Information Systems

604

this website and had their prices searched on the page

of Mouser (Electronics, 2017) and Amazon (Amazon,

2017). The types of sensors supported refer to the

number of sensor types that can be used in the device,

that is, how many sensors with different purposes are

supported, allowing a wider range of applications for

the device.

Table 1: Comparative analysis of related works.

Platform Cost

# Supported

Sensors

Work 1/2

US$3.85 4

Work 3 US$10.26 7

Work 4 - 7

Work 5 US$6.21 -

Work 6 US$49.55 4

Work 7 US$2.50 3

Work 8 US$2.50 2

Proposed US$1.78 8

The project developed in this work has as one

of its objectives to be low cost. The proposed plat-

form has a cost of US$1.78 (AliExpress, 2017f) and

in addition to the ESP8266 Wi-Fi module, at a cost

of US$2.21 (AliExpress, 2017d), the total cost of the

device is US$3.99, without considering the sensors

and actuators that could be used. The sensors used

in the experiments were: luminosity (LDR), which

costs US$0.40 (AliExpress, 2017g); temperature and

humidity (DHT11) at a cost of US$0.78 (AliExpress,

2017b); and movement (PIR), which has a cost of

US$0.78 (AliExpress, 2017e). For the actuator, a one-

channel relay module was used that costs US$0.53

(AliExpress, 2017a). In addition to these three sen-

sors used, the device can provide support for ﬁve

different sensors, which are: infrared (E18-D80NK)

(UsinaInfo, 2017), sound intensity (KY-038) (Filipe-

Flop, 2017d), rainfall detection (YL-83) (FilipeFlop,

2017a), soil moisture (YL-69) (Makers, 2017) and gas

detection (MQ-2) (FilipeFlop, 2017b). Considering

sensors/actuators that use 1 data pin, such as those

supported by the device, up to 20 components could

be used in a single device (Arduino, 2017).

5 CONCLUSION

Most of the existing residential automation solutions

are complex and they have high cost (Freitas, 2016).

The purpose of this work is the development of a low-

cost, modular residential automation solution that can

be easily adopted.

The use of Home Assistant application provides

the control and monitoring of the equipment through

the Internet. This application, together with the use of

WiFi communication, has increased the scalability of

the project. For communication between devices with

the Home Assistant, the MQTT protocol was used,

which reduced the hardware requirements and faci-

litated its implementation due to its low complexity.

When compared to the related works, the proposed

project has the lowest cost platform and supports a

greater variety of sensors. Its installation, conﬁgura-

tion and maintenance is straightforward, which faci-

litates its use. In addition, this platform allows the

expansion to control more equipment.

Future work includes the implementation of new

sensors and actuators for the device, since only three

types of sensors have been tested. Automation

can also be made in the Home Assistant (Assistant,

2017a) application, creating automatic rules for con-

trolling electronic equipment, such as turning on lig-

hts when it is dark and detecting presence. Another

future work would be to use the device data for big

data applications. This data can be stored and used by

some software with artiﬁcial intelligence, such as pat-

tern recognition, making the devices predict actions

that users would perform.

REFERENCES

AliExpress (2017a). 1 pcs 5v low level trigger one 1 channel

relay module interface board shield for pic avr dsp arm

mcu arduino. Available at: https://goo.gl/PTzNiQ.

[Accessed 21 Sep. 2017].

AliExpress (2017b). 1x dht11 dht-11 digital temperature

and humidity temperature sensor for arduino diy kit.

Available at: https://goo.gl/zcN4i9. [Accessed 21 Sep.

2017].

AliExpress (2017c). Aliexpress. Available at: https://pt.

aliexpress.com. [Accessed 13 Oct. 2017].

AliExpress (2017d). Diy esp8266 esp-01s wireless wiﬁ

sensor for arduino esp-01 advanced version wire-

less transceiver. Available at: https://goo.gl/TJQKxS.

[Accessed 21 Sep. 2017].

AliExpress (2017e). Hc-sr501 adjust infrared ir pyroe-

lectric infrared pir module motion sensor detector

module we are the manufacturer. Available at:

https://goo.gl/N7u7aZ. [Accessed 21 Sep. 2017].

AliExpress (2017f). Nano 3.0 controller compatible for ar-

duino nano ch340 usb driver no cable. Available at:

https://goo.gl/pnVfcn. [Accessed 13 Oct. 2017].

AliExpress (2017g). Photosensitive sensor module light

detection module for arduino. Available at: https://

goo.gl/gDTjB9. [Accessed 13 Oct. 2017].

Amazon (2017). Amazon. Available at: https://

www.amazon.com. [Accessed 13 Oct. 2017].

Arduino (2017). Arduino nano. Available at: https://goo.gl/

TQB9En. [Accessed 07 Oct. 2017].

Modular Automation Design for Equipment Management

605

Assistant, H. (2017a). Automating home assistant. Avai-

lable at: https://goo.gl/2RY6FW. [Accessed 17 Oct.

2017].

Assistant, H. (2017b). Awaken your home. Available at:

https://home-assistant.io/. [Accessed 27 Apr. 2017].

Barros, M. (2015). Mqtt - protocolos para iot. Available at:

https://goo.gl/zVdBSD. [Accessed 24 Apr. 2017].

Braga, I. N. C. (2017a). A internet das coisas (iot) (art2054).

Available at: https://goo.gl/YhoqLA. [Accessed 14

Apr. 2017].

Braga, I. N. C. (2017b). A onda do arduno (col001). Avai-

lable at: https://goo.gl/FqfQLS. [Accessed 19 Sep.

2017].

Chi, Q., Yan, H., Zhang, C., Pang, Z., and Xu, L. D. (2014).

A reconﬁgurable smart sensor interface for industrial

wsn in iot environment. IEEE Transactions on Indus-

trial Informatics, 10(2):1417–1425.

e Cia, A. (2013). Sensor de umidade e temperatura dht11.

Available at: https://goo.gl/XZto8d. [Accessed 21

Sep. 2017].

e Cia, A. (2014). Sensor de presena com mdulo pir dyp-

me003. Available at: https://goo.gl/ut8hTc. [Accessed

07 Oct. 2017].

Eclipse (2017). Mosquitto. Available at: https://

mosquitto.org/. [Accessed 23 Sep. 2017].

Electronics, M. (2017). Mouser electronics. Available at:

http://br.mouser.com. [Accessed 13 Oct. 2017].

FilipeFlop (2017a). Sensor de chuva. Available at: https://

goo.gl/cGJgL7. [Accessed 16 Oct. 2017].

FilipeFlop (2017b). Sensor de gas mq-2 inﬂamvel e fu-

maa. Available at: https://goo.gl/H2XycT. [Accessed

16 Oct. 2017].

FilipeFlop (2017c). Sensor de luminosidade ldr (light de-

pendent resistor) 5mm. Available at: https://goo.gl/

q2uoa7. [Accessed 21 Sep. 2017].

FilipeFlop (2017d). Sensor de som ky-038 microfone. Avai-

lable at: https://goo.gl/g37Hyf. [Accessed 16 Oct.

2017].

Foundation, P. S. (2017). Common gateway inter-

face support. Available at: https://docs.python.org/

2/library/cgi.html. [Accessed 24 Sep. 2017].

FreeRTOS (2017). Freertos faq - memory usage, boot times

& context switch times. Available at: https://goo.gl/

43E8LA. [Accessed 25 Feb. 2017].

Freitas, L. (2016). O que ainda falta para a internet das

coisas decolar. Available at: https://goo.gl/HK01y1.

[Accessed 25 Apr. 2017].

Higuera, J., Hertog, W., Perlvarez, M., Polo, J.,

and Carreras, J. (2015). Smart lighting system

iso/iec/ieee 21451 compatible. IEEE Sensors Journal,

15(5):2595–2602.

Kelly, S. D. T., Suryadevara, N. K., and Mukhopadhyay,

S. C. (2013). Towards the implementation of iot for

environmental condition monitoring in homes. IEEE

Sensors Journal, 13(10):3846–3853.

Kruger, C. P., Abu-Mahfouz, A. M., and Isaac, S. J. (2013).

Modulo: A modular sensor network node optimised

for research and product development. In 2013 IST-

Africa Conference Exhibition, pages 1–9.

Magno, M., Polonelli, T., Benini, L., and Popovici, E.

(2015). A low cost, highly scalable wireless sensor

network solution to achieve smart led light control for

green buildings. IEEE Sensors Journal, 15(5):2963–

2973.

Makers, M. (2017). Sensor de umidade de solo yl-69. Avai-

lable at: https://goo.gl/wJKFn1. [Accessed 16 Oct.

2017].

Mikhaylov, K. and Huttunen, M. (2014). Modular wireless

sensor and actuator network nodes with plug-and-play

module connection. In IEEE SENSORS 2014 Procee-

dings, pages 470–473.

Mikhaylov, K. and Paatelma, A. (2015). Enabling modular

plug amp;play wireless sensor and actuator network

nodes: Software architecture. In 2015 IEEE SEN-

SORS, pages 1–4.

Muratori, J. R. and B, P. H. D. (2017). Captulo i. Available

at: https://goo.gl/5h3JHZ. [Accessed 25 Apr. 2017].

Song, E. Y. and Lee, K. (2008). Understanding ieee 1451-

networked smart transducer interface standard - what

is a smart transducer? IEEE Instrumentation Measu-

rement Magazine, 11(2):11–17.

Suh, C. and Ko, Y. B. (2008). Design and implementation of

intelligent home control systems based on active sen-

sor networks. IEEE Transactions on Consumer Elec-

tronics, 54(3):1177–1184.

Takahashi, P. (2016). Internet das coisas: o que falta

para decolar? Available at: https://goo.gl/4KTQaH.

[Accessed 25 Apr. 2017].

Thomsen, A. (2013). Controlando lmpadas com mdulo rel

arduino. Available at: https://goo.gl/brNb6p. [Acces-

sed 07 Oct. 2017].

Thomsen, A. (2015). Tutorial mdulo wireless esp8266 com

arduino. Available at: https://goo.gl/o0Todn. [Acces-

sed 21 Apr. 2017].

Thomsen, A. (2017). Qual mdulo esp8266 comprar? Avai-

lable at: https://goo.gl/a6P8ZY. [Accessed 07 Oct.

2017].

Tzapu (2017). Wiﬁmanager. Available at: https://

github.com/tzapu/WiFiManager. [Accessed 09 Oct.

2017].

UsinaInfo (2017). Sensor de proximidade infravermelho -

deteco 3 a 80cm - e18-d80nk. Available at: https://

goo.gl/bKHwsn. [Accessed 16 Oct. 2017].

Wang, J. (2013). Zigbee light link and its applicationss.

IEEE Wireless Communications, 20(4):6–7.

Zambarda, P. (2014). Captulo i. Available at: https://goo.gl/

4S3M6G. [Accessed 25 Apr. 2017].

ICEIS 2018 - 20th International Conference on Enterprise Information Systems

606