Fall Detection Using Intelligent Walking-Aids and Machine Learning Methods

Walking aids are commonly given to older adults to prevent falls, but paradoxically, their use has been identified as a risk factor for falling, which is a prevalent issue among this population, causing serious injuries, disabilities, and even death. This has resulted in a significant increase in public health care and the development of remote health monitoring technology to enhance home care devices. One of the key issues being addressed is the identification of falling incidents, which can aid in the rapid arrival of assistance and prevent additional harm. This paper is to develop intelligent walking aids using machine learning methods and an M5Stack Core2 microcontroller. These aids analyze information from various sensors such as accelerometers and gyroscopes to identify falls. Touch and location sensors are also utilized to determine the device’s usage and location in case of an emergency. The collected data is sent to a web server in JSON format via the M5Core2’s WiFi module, allowing for a quick response if necessary. The fall detection system has been extensively tested, resulting in 99.62% accurate identification of falls.