Gesture Based Home Automation Simulation Using Machine Learning

Home automation is the process of controlling various aspects of a smart home, such as lighting, security, and entertainment, through a centralized system. The proposed system is a novel approach to home automation using gestures, which are captured by a camera and processed by a classification algorithm. Currently proximity sensors and Wi-Fi methods are used to control which are the best ways of automating, but it has some drawbacks. Alexa or other such devices even mobile phones could be used to control the electrical appliances. Whereas proximity sensors which may automatically control the appliances without our willingness. This may increase the power expenditure. The problems are solved by the means of gestures which does not require internet and does not control without the willingness as machine learning is employed. The design is a set of gestures that can be used to perform common tasks, such as turning on/off lights, adjusting the thermostat, playing music, and locking/unlocking doors. The simulated home automation system utilizes Arduino Uno, keypad, relay module and a lamp as load for effective appliance control. The performance is evaluated and usability of our system through experiments and user studies. The simulation replaces the gesture recognition with the keypad, will turn on the lamp if it reads 1 and turn off if it reads 0. The system can achieve high accuracy and user satisfaction and demonstrate the potential of gesture-based home automation.