Mobile Robot for Object Detection using an IoT System

摘要

The Internet of Things (IoT) is a technological revolutionary paradigm which has enabled communication with a wide number of devices through the Internet. In this paper we present the design of a mobile robot which uses an Internet of Things architecture to be controlled remotely. This robot includes a camera which is used to obtain landscape images from the mobile robot. A web server allows to store the commands to control the movements of the mobile robot in a MySQL database as well as the transmission of commands between the server and the mobile robot; by using web pages with PHP and an Android APP. Our electronic system is based on WiFi Module ESP8266 which offers advantages to incorporate open software architectures. The landscape images are processed in a computer using python and machine learning libraries to identify interesting objects on the images. The machine learning algorithms recognise on the landscape images two type of objects classes: agave plants and stones. We use a convolutional neural network for the object detection known as You Only Look Once (YOLO) which achieves 81.2% of accuracy for recognising both classes.