Developing a Machine Learning Algorithm for Service Robots in Industrial Applications

Abstract

Robots, which have mostly been effective in areas such as industrial, agricultural, and production facilities, have started to take a place in the service sector, as their technologies have become lower in cost and more easily accessible. This situation has attracted the attention of companies and researchers and has accelerated studies on this subject. In this study, an algorithm was developed for the autonomous mobile robot to serve in industrial areas. In line with this study, it was ensured that the autonomous mobile robot mapped the working environment, determined the working station in this environment, and then carried out transport operations between these working stations in accordance with a given work order. After the mobile robot fulfilled the work order, it went into a waiting state until a new work order was received. For the mobile robot to save energy, it was ensured that it waited close to the point where the work order came in the most, by means of machine learning in the waiting position. The developed algorithms were designed using the NI LabVIEW environment and then simulated in the RobotinoSIM environment and physically tested using the Robotino autonomous mobile robot platform. The experimental results showed that mapping and location reporting using an RGB camera, odometry, and a QR code eliminated permanent location errors, and the robot completed 50 work orders with 100% accuracy.