Image Processing Based Human Pursuing Robot

Abstract

In recent years there has been increase in development of human pursuing robots which can be used as daily life support robots. The primary goal is to design and fabricate a robot that not only tracks the target but also moves according to it. For implementing this project, a barcode was used as target that robot needs to follow. OpenCV provides an interface to capture live stream with camera. In order to detect the barcode, a program is written in python which is interfaced with OpenCV library. After capturing a video from the camera, it converts it into gray scale video and display it frame by-frame. After detecting the barcode from the video frame, black and white lines of an image is detected and the centroid will be calculated. Based on the position of the centroid, commands are given to the robot to move accordingly. Ultrasonic sensor is used to avoid collision between the robot and obstacles. As a result, the robot pursues the target and it can be used as an assisting system for handicapped people for carrying luggage or can be used in airports, railway stations as a luggage carrier.