A Mobile Robot Based on Edge AI

Abstract

Recent advancements in technology have enabled the emergence of Industry 4.0, this term is used to describe the integration of various advanced technologies such as artificial intelligence and robotics in industrial settings. The presence of defective products during production incurs additional costs, and traditional manual methods of equipment inspection prove to be inefficient in such a dynamic environment. In this study, we introduce a robot designed specifically for this scenario, capable of performing tasks that require autonomous movement to specific areas of an industrial plant. To achieve this, we employ the concept of Edge AI, applying artificial intelligence on a localized edge computing device. The robot utilizes computer vision through the state-of-the-art YOLOv7 CNN and incorporates feedback control to facilitate its locomotion. The hardware components of this robot include a Jetson Xavier NX, Raspberry Pi 4, a camera, and a LIDAR. Additionally, we conducted a comprehensive performance analysis of the object detection method, measuring metrics such as frames per second (FPS), CPU and GPU consumption, and RAM usage.