Gestures detection and device control in AAL environments using machine learning and BLEs

Abstract

This paper presents a system for detecting gestures and controlling devices in Ambient Assisted Living (AAL) environments using machine learning and Bluetooth Low Energy (BLE) technology. The system consists of two main components: a device equipped with a set of sensors to detect hand gestures via IMU sensor and a BLE-enabled hub that receives the gesture data and controls the lighting of the house. The hub uses machine learning algorithms to recognize hand gestures and transmit the corresponding commands to the devices. The hub, in turn, uses wifi to communicate with the devices and execute the appropriate actions based on the received commands. The proposed system's performance evaluation was carried out through a series of experiments in a AAL environment. The results demonstrate that the system is capable of accurately detecting hand gestures and controlling various devices such as lights, where the model's performance yields successful predictions with an accuracy rate of 90%. The proposed system provides a user-friendly and intuitive way for elderly or people with disabilities to control their environment without the need for complex interfaces or physical buttons. Furthermore, the system can be easily extended to support more gestures and devices, making it a flexible and scalable solution for AAL environments.