Saturday, 3 May 2014

Gesture recognition system to operate household appliances as an aid to handicapped and elderly people

Hi, guys.. I am back again with a new project. As a part of my Electrical group project me and two of my friends.. Ashutosh and Girish have developed a Gesture recognition system to remotely operate household appliances. The prototype has been tested and could act as an aid to handicapped and elderly people (bed ridden people).

Find and download all the codes from here {please like and comment}

The complete set up

Objective: Designing a system which can act as an aid to the handicapped and/or elderly people such that they can operate the household appliances by making a gesture rather than physically operating them.

Scope: In this project, we have developed a simple prototype of a gesture recognition system which understands specific gestures and acts to operate an appliance associated with that action. This system is designed for handicapped or elderly people, therefore there is an action confirmation mechanism enabled in the system so that the appliances are not operated by accident. We demonstrate handling of an LED – simulating Light Bulb or a CFL lamp, a 3V motor – simulating fan and a buzzer – simulating analarm that can be raised in the case of emergency.

Design: System is divided into two parts: a hand module and a base module. Hand module will be worn by the user and base module will be the one that operates the various appliances in the house. Hand module consists of an accelerometer (MPU6050),an Arduino Uno board and an 433MHz transmitter. The base module has a 433 MHz receiver interfaced to another Arduino Uno board along with a 3V motor, an LED and a buzzer which simulate – fan, Bulb or a CFL and Alarm respectively. The InvenSense's MPU-6050 sensor contains a MEMS accelerometer and a MEMS gyro in a single chip. It is very accurate, as it contains 16-bits analog to digital conversion hardware for each channel. Therefor it captures the x, y, and z channel at the same time. It acts as slave with Arduino and communicates data through SDA and SCL pins. Arduino Uno – This board has 14 digital I/O pins and 6 analog input pins. Board can be powered via USB or 5V adapter connected to AC mains.

Hand Module configuration: 

Accelerometer pin SCL
Analog Pin A5 (input)
Accelerometer pin SDA
Analog Pin A4 (input)
433 MHz Transmitter data pin
Digital Pin 12 (output)

Base Module configuration:

Digital Pin 13 (output)
3V motor
Digital Pin 6 (output)
Digital Pin 12 (output)
433 MHz Receiver data pin
Digital Pin 11 (input)


We have five mutually orthogonal gestures for
1.  Turning lights ON (Hand held vertically up)
2.  Turning lights OFF (Hand held vertically down)
3.  Starting/ increasing speed of fan (rotating hand clockwise along forearm)
4.  Stopping/decreasing speed of fan (rotating hand anti-clockwise along forearm)
5.  Alarm (placing hand on heart)

These easy gestures were chosen keeping in mind that user is likely to be a bed-ridden person. From the gestures chosen, we can see that gestures 3 and 4 could be done accidentally by a person. Therefore, we have an action confirmation gesture such that fan will be increased (or decreased) only when gesture 3 (or 4) will be followed by action confirmation gesture. For simplicity, action confirmation gesture is nothing but gesture 1 from the list. A simple loop function keeps on monitoring the inputs received from Accelerometer and every orientation has certain accelerometer values which can be detected and corresponding signals are sent to the base module. Base module operates the appliances based on the signals it receives from the hand module.

Block Diagram:  

Block Diagram

Future Development:

1.  More appliances could be easily added by assigning new gestures to them.
2.  Although this prototype was designed for handicapped and elderly people, the small scale wearable device with gesture recognition technology could be used as a universal appliance controller for any person.
3.  The gestures were chosen, keeping in mind that the system was to be used for handicapped and elderly people, if it is to be used as a wearable technology, more generic gestures could be chosen to provide easiness.
4.  The 433 MHz RF pair provides no security in the communication with each other. For wearable technology module this feature can be added in the system.

The Base Module {sorry guys couldn't create a video - was too busy with end semester exams}

The Hand Module

Base Module received gestures

Find and download all the codes from here {please like and comment}