Drowsiness detection in drivers with a smartwatch Sonia D az-Santos1 Pino Caballero-Gil2

2025-05-03 0 0 776.42KB 12 页 10玖币

侵权投诉

Drowsiness detection in drivers with a

smartwatch

Sonia D´ıaz-Santos1, Pino Caballero-Gil2

University of La Laguna, St. San Francisco de Paula, 19, 38200, San Crist´obal de La

Laguna, Tenerife, Spain

sdiazsan@ull.edu.es, pcaballe@ull.edu.es

Abstract. The main objective of this work is to detect early if a driver

shows symptoms of sleepiness that indicate that he/she is falling asleep

and, in that case, generate an alert to wake him/her up. To solve this

problem, an application has been designed that collects various parame-

ters, through a smartwatch while driving. First, the application detects

the driving action. Then, it collects information about the most signiﬁ-

cant physiological variables of a person while driving. On the other hand,

given the high level of sensitivity of the data managed in the designed

application, in this work special attention has been paid to the secu-

rity of the implementation. The proposed solution improves road safety,

reducing the number of accidents caused by drowsiness while driving.

Keywords: road safety, smartwatch, drowsiness, physiological variables,

safe driving, safe scheduling

1 Introduction

Traﬃc accidents often endanger the lives not only of the driver but also of other

people on the road. It is therefore necessary to do everything possible to reduce

their number. Among the options to achieve this, this work has opted for the

development of innovative technologies to address the problem. Speciﬁcally, the

present study has focused on identifying physiological variables that character-

ize sleepiness or fatigue while driving, with the aim of using them to reduce

the number of accidents caused by that reason, since, in general, drowsiness is

involved, directly or indirectly, in 15-30% of traﬃc accidents.

In this work, a study of the most relevant physiological variables that allow us

to conclude whether a person is falling asleep has been carried out. As discussed

in this paper, according to various publications, these variables are: heart rate,

electrocardiogram, respiratory function and stress. Speciﬁcally, by means of the

smartwatch used in this work, it has been possible to collect the data of some of

these physiological variables of a person, to locate the signs of drowsiness and

check, in real time, if the person is falling asleep [1].

On the other hand, to detect the action of driving, the smartwatch has a set

of sensors such as accelerometer, gyroscope, pedometer and Global Positioning

System (GPS).

arXiv:2210.04066v1 [cs.CR] 8 Oct 2022

2 Sonia D´ıaz-Santos, Pino Caballero-Gil

Thus, this work starts from the intersection between the list of physiological

variables descriptive of sleep and that of the sensors available in the used smart-

watch. In particular, some parameters such as Heart Rate Variability (HRV) have

been used. Then, these collected data have been analyzed to detect drowsiness

in drivers. For this purpose, diﬀerent sensors such as PhotoPlethysmoGraphy

(PPG) and ElectroCardioGram (ECG) sensors have been used. The PPG sensor

uses a light-based technology to detect the rate of blood ﬂow controlled by the

pumping action of the heart. In addition, other sensors such as the accelerome-

ter, gyroscope, pedometer and GPS, integrated into the watch, have been used

to monitor the user’s physical activity [2].

Other factors have also been taken into account, such as the time of day,

allowing for the circadian rhythm according to which physical, mental and be-

havioral changes occur in 24-hour cycles. In fact, it is known that fatigue-related

accidents are strongly dependent on the time of day, as well as on the type of

road, especially on monotonous roads. In addition, other factors to consider are

a person’s age, gender and regular physical exercise, to determine the normal

parameters for each individual. For example, heart rate in athletes is usually

lower than in sedentary people, and heart rate is usually lower in women than

in men. Keeping all these indicators in mind, a better analysis of the drowsiness

data has been possible [3].

This document is structured as follows. Section II provides data from the

used smartwatch and platform. Section III contains a discussion of physiological

variables and sensors of interest for this work. Section IV provides a description

of the design and functionalities of the proposed application. Section V describes

some features of the secure implementation of the application. Finally, section

VI closes the paper with some conclusions and future work.

2 Smartwatch and platforms used

This section provides some details of the smartwatch chosen to measure the

driver data, as well as of the platform used to develop the applications [4].

2.1 Hardware

For the development of this work, the smartwatch Samsung Galaxy Watch 4 LTE

has been used. This watch has the Wear OS Powered by Samsung, which allows

health monitoring 24 hours a day. It has a sensor BioActive that measures ECG

and blood pressure in real time. To measure blood pressure, it uses an optical

PPG heart rate sensor, and to measure ECG, it uses an electrical heart sensor.

It also allows measuring body composition through the Bioelectrical Impedance

Analysis (BIA) sensor. This device allows blood oxygen and stress levels to be

measured to obtain a complete sleep analysis. It has two motion sensors, the

accelerometer and the gyroscope, which allow to know the location with the

GPS sensor, and calculates the steps with the pedometer sensor. In terms of

connections, it has Bluetooth 5.0 and Wi-Fi connection [5].

Drowsiness detection 3

2.2 Software

This section deﬁnes the software used by the devices, including the communi-

cation architecture between the systems, as well as the technologies and devel-

opment environments. It also mentions the Health Platform and the Wear OS

community.

2.2.1 Architecture communication scheme Bluetooth or Wi-Fi connec-

tions are required for communication between the diﬀerent devices, as shown in

Figure 1. The smartwatch with the Wear OS Powered by Samsung has a Blue-

tooth connection to connect to the cell phone. The cell phone has the application

Samsung Health, which collects the necessary data of physiological variables, as

well as the application Samsung Health Monitor to obtain the data of the blood

pressure and electrocardiogram of the person using the watch. The application

is created on the computer and installed on the smartwatch through the devel-

opment environment Android Studio. The application is installed on the watch

through the aforementioned connections. In addition, with One UI Watch, com-

patible apps are automatically installed on the watch when they are downloaded

to the cell phone.

Fig. 1. Communication scheme

文档加载中……请稍候！
如果长时间未打开，您也可以点击刷新试试。

下载文档到电脑，查找使用更方便

10 玖币 0人已下载

立即下载

摘要：

DrowsinessdetectionindriverswithasmartwatchSoniaDaz-Santos1,PinoCaballero-Gil2UniversityofLaLaguna,St.SanFranciscodePaula,19,38200,SanCristobaldeLaLaguna,Tenerife,Spainsdiazsan@ull.edu.es,pcaballe@ull.edu.esAbstract.Themainobjectiveofthisworkistodetectearlyifadrivershowssymptomsofsleepinessthatin...

展开>> 收起<<

Drowsiness detection in drivers with a smartwatch Sonia D az-Santos1 Pino Caballero-Gil2.pdf

共12页,预览3页

还剩页未读，继续阅读

声明：本站为文档C2C交易模式，即用户上传的文档直接被用户下载，本站只是中间服务平台，本站所有文档下载所得的收益归上传人(含作者)所有。玖贝云文库仅提供信息存储空间，仅对用户上传内容的表现方式做保护处理，对上载内容本身不做任何修改或编辑。若文档所含内容侵犯了您的版权或隐私，请立即通知玖贝云文库，我们立即给予删除！

Drowsiness detection in drivers with a smartwatch Sonia D az-Santos1 Pino Caballero-Gil2

相关推荐

开通VIP享超值会员特权

作者详情

相关内容

热门标签

举报选择: