AN OMNI-DIRECTIONAL MOBILE ROBOT USING DEEP LEARNING AND INTERNET OF THINGS FOR AGRICULTURAL WAREHOUSES APPLICATION
Febriyanti, Abdul Kadir Muhammad, Imran Habriansyah and Indarwati

Center for Mechatronics and Control System, Mechanical Engineering Department, State Polytechnic of Ujung Pandang, Makassar, Indonesia.

Abstract

The purposes of this study are to develop an active caster robot for agricultural warehouses application, to develop deep learning - based computational codes and to perform laboratory scale experiments. The system used in this paper consist of four dc motors as actuators, four ultrasonic sensors to keep the robot from crashing into shelves or other objects, a box used as a storage container for goods with an ultrasonic sensor to detect the presence of goods, and a camera. Deep learning method was used to keep the robot on the track. An internet of things - based control system was designed to control the robot. Performances of the robot were examined through laboratory scale experiments. The results of the conducted experiments are presented and discussed.

Keywords: deep learning, internet of things, omni-directional mobile robot.

Topic: Agriculture engineering