Intelligent Gesture Controlled Automation

Abstract

The project focuses on simplifying various tasks by enabling individuals to control appliances in their surroundings wirelessly by using simple hand gestures. An alternative to pushing buttons on remotes and switches, this method offers various advantages over its counterparts such as:- 1) Allowing quick and convenient method of controlling a large number of appliances 2) Empowering those with loss of vision, partial motor skills, or mobility 3) Efficient centralization of control and monitoring The device used to achieve this will consist of an accelerometer, a microcontroller and an RF transmitter that will be attached to the hand of the user. The respective devices that are to be controlled will be connected to a receiver and multiple decoders. Simple hand gestures will be analyzed by the microcontroller and the corresponding instruction will be invoked from a look up table and transmitted digitally. Accordingly at the receiver end, the respective appliance will be controlled as desired by the user. A machine learning algorithm known as Learning Vector Quantization (LVQ) is implemented in this project to create clusters associated with different hand gestures which are determined by the user. As a result, different users may set different gestures for control as per their convenience. The application scope of this project may range from home automation to more important things such as facilitating handicapped individuals, military drone control, sign language translation, security systems and medicinal automation.