ROBOTIC SHELF STOCK MONITORING SYSTEM

byDnyanesh DeoFeroz ChaodhriRaja Chowdhury

project guideMr. Gopalkrishnan Naraynan

The Problem

• Reduced sales despite that product being in the stock.

• Unnecessary stock piles up even before the current stock has depleted.

• Problem in the area of big parking lots.

Our Application

• An autonomous robot that will roam in the mall along a line.

• Monitor shelves for stock content through image processing in LABVIEW.

• For empty or near empty shelves, it will send a message to the control room to fill the stock.

• To analyze empty slots for available parking slots.

Project Description

• Application consists of a line follower robot.

• USB camera is mounted on two bars

• Robot is made using a aluminium frame.

• Line follower will handle the horizontal movement of the robot or camera.

• USB camera will give the view of the shelf.

• Camera is mounted on a pulley system.

• Pulley system consists of simple motor and support from bars.

• Pulley system will handle the vertical movement of the robot or camera.

• Photoelectric Proximity sensors for crowd detection.

• Images acquired by USB camera will be processed using LabVIEW

Mechanical Model

Camera mounting

Supporting platform

Base

Motor

Line Follower

• Follows black line on white surface.

• Four wheel drive.

• Speed control by motor driver.

• Hardware for Line follower Robot-

1. Arduino Board.2. 12V DC motors.3. Wheels.4. IR Sensors.5. H-Bridge Motor Drivers.6.12V lead acid battery

H - Bridge

An H bridge is an electronic circuit that enables a voltage to be applied across a load in either direction. These circuits are often used in robotics and other applications to allow DC motors to run forwards and backwards.

Other hardware required is -1. USB Camera for image acquisition

2. Motor in pulley system

3. Photoelectric Proximity Switch

Software needed -

● Arduino IDE● LabVIEW

What We Have Tried§ H – Bridge IC L298§ H – Bridge IC A3940KLWT with Power Mosfet§ H – Bridge IC MC33886 (Working)§ IR Sensor (Working but NO proper mounting)§ Photo-electric Sensor§ Photo-electric Switch (Working)§ RF module for wireless communication (Working only one way)

Hurdles§ SOLVED1. Camera Mounting2. Finding appropriate Motor Driver3. Photo-electric Switch for Line Following

§ UNSOLVED1. Compact chassis (For proper Line Following)2. Tire Grip3. Four Wheel Drive Or Two Wheel Drive4. Battery Life

LabVIEW

• Vision Acquisition and Vision Assistant Module.

• OCR/OCV pattern recognition feature.

• Symbols detection in region of interest.

• Status of shelf will be calculated on the basis of the number of symbols detected.

● For communication between robot and control room.

● TCP/IP communication using UDP Multicast.● Hosting LAN● In case of emergencies StockEye can be

controlled from Control Room

What We Have Tried§ IMAGE PROCESSING1. Edge Detection2. Pattern Matching3. Thresholding, Segmentation4. OCR/OCV (Working)

§ Line Following1. Hough Transform2. Edge Detection

§ COMMUNICATION

1. NI VISA Serial Module2. Arduino LIFA Module3. TCP/IP VI (by hosting LAN)

Hurdles§ SOLVED1. Image Processing2. Communication (From StockEye to Control Room)

§ UNSOLVED1. Color Image Processing (Appropriate Camera with Auto Focus)2. Two way communication3. Distance of Camera from Shelf