International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 12 | Dec 2025
P-ISSN: 2395-0072
www.irjet.net
IoT-Based Smart Walking Assistant for Fall Detection in the Elderly Hemashree H C1, Adil Ahmed2, Sumanth P Bellad 3, Vijayalakshmi S4, Spoorthi G R5 1Assistant Professor, Information Science and Engineering, Bapuji Institute of Engineering and Technology,
Davangere, affiliated to VTU Belagavi, Karnataka, India.
2,3,4,5,Bachelor of Engineering, Information Science and Engineering, Bapuji Institute of Engineering and
Technology, Davangere, affiliated to VTU Belagavi, Karnataka, India. ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Falls are a major health risk for elderly individuals, leading to injuries, hospitalization, and reduced independence. Traditional walkers provide physical support but lack intelligence for fall monitoring and alerting. This work proposes an IoT-enabled smart walking assistant equipped with accelerometer– gyroscope sensors for fall detection, along with heart rate, SpO₂, body temperature, obstacle detection, and live caregiver alerts. The system uses ESP32 and LoRa communication to support real-time tracking and emergency notification, aiming to reduce injury severity and improve elderly safety. Key Words: Smart walking assistant, elderly safety, internet of things, fall prevention, physiological sensors.
1.INTRODUCTION The global elderly population is growing rapidly, with the number of individuals aged 60 and above projected to reach 2.1 billion by 2050 [1]. While increased longevity is a positive societal achievement, it also presents significant challenges, especially in the realm of healthcare. Among these challenges, falls represent one of the most severe risks for the elderly. Statistics show that one in ten falls leads to an injury that causes older adults to limit their activities for at least a day or seek medical attention [2]. The global elderly population (60+ years) is expected to reach 2.1 billion by 2050, increasing the risk and impact of falls. One in ten falls results in injury requiring medical support. Wearable and ambient systems have helped, but discomfort, privacy issues, and indoor limitations restrict usability. IoT-based health solutions enable continuous remote monitoring and timely alerts. This paper presents a smart walking assistant that improves mobility, detects Falls instantly, and notifies caregivers through a mobile app.
2. LITERATURE REVIEW The rapid advancement of fall detection technologies has led to a wide array of methods aimed at enhancing elderly safety and reducing healthcare burdens. The literature generally categorizes fall detection methods into many types [3], [4], [5]. Each method offers unique benefits and faces specific challenges, as discussed below. Fall-detection systems employ several methods, each with distinct advantages and limitations. Wearable sensor–based approaches offer high accuracy and enable precise motion tracking, but they often face challenges related to user compliance. Ambient sensor systems eliminate the need for wearable devices, making them more comfortable for users; however, they are typically limited to indoor environments and raise privacy concerns. Vision-based methods provide very accurate pose estimation and fall recognition, yet they require costly camera setups and also involve significant privacy issues. More recently, machine learning combined with IoT has enabled intelligent fall prediction and faster emergency response by automatically sending alerts to caregivers, although this approach relies heavily on cloud infrastructure and large volumes of data. While machine learning significantly improves motion classification accuracy, its performance strongly depends on the diversity and quality of the training dataset.
2.1 Wearable sensor- based fall detection. Wearable sensors such as accelerometers and gyroscopes are the most commonly used tools for fall detection due to Their portability and ability to capture real-time motion data [6]–[11]. Studies show that combining these sensors enables accurate discrimination between falls and normal movements. For example, Wu et al. [12] achieved high accuracy using quaternion-based motion analysis, while Siregar et al. [13] reported 93.75% accuracy with an Arduino-based alerting system. Ahn et al. [14] demonstrated 100% sensitivity in pre-impact fall detection using angular velocity and trunk inclination.
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 131