Design and implementation of autonomous robot with solid object identification algorithm

Abstract

Object tracking and identifying of solid shapes are the key issues in autonomous mobile robot navigation. In this paper, we present a successful design and implementation approach of an effective algorithm for mobile robot object tracking and identification. The robot is capable to identify pre-defined solid objects which are kept in any arbitrary location. Unlike some other object detection algorithms, in this approach the colour of the objects will not be considered in completing the tasks, which makes the robot more reliable in various light conditions and ambient conditions. The algorithm developed can also handle different field condition to avoid obstacles. The robot navigation system is based on low cost and reliable proximity sensor network. This approach can be effectively used in industrial applications such as product line automation, defective part removing processes, automated vehicle driving systems and battlefield mine detection.