Development of a High Precision UWB/Vision-based AGV and Control System

Abstract

AGVs play an important role in unmanned factories and have broad prospects for development. Indoor vehicles are usually guided by magnetic tape / magnet, inertia, lidar, etc. These methods have some disadvantages. For example, magnet method is inconvenient due to a regular repair or replacement of tapes. In this paper, an advanced UWB method is used to navigate the AGV to the vicinity of the working position, and then computer vision technology is used to accurately determine the position of the vehicle, thereby establishing a high-precision AGV self-localization method. UWB can be applied to the situation of multi-AGVs interaction. Depth camera can also be used for obstacle detection and workpiece detection. Meanwhile, in order to make all different hardware and software work properly, an easy-scalable AGV control system is designed. Simple sensors can achieve complex functions.