Vision-Based Navigation Of An Autonomous Guided Vehicle For Tracking

This project proposes a navigation control system for an Autonomous Guided Vehicle (AGV) by detecting and recognizing line tracking with USB (Universal Serial Bus) camera. In this project a commercial robot kit, a laptop computer and a USB camera are used as the main components of the system. The use of Microsoft DirectShow technique and COM programming, the USB camera will enable the system to obtain ' the digital images directly without using frame grabbers. This feature makes the system more economical and compact. The vision-based navigation system structure is composed of several processes such as grabbing images, track detection, fuzzy logic controller and' motor drive controller. During the navigation process, the AGV can recognize the straight and crossing track lines, detect the obstacle that might appear in the navigational path, calculate the position and orientation of the AGV. When the robot has confirmed its position and orientation, the fuzzy controller is used to keep the AGV on the track. This proposed navigation system only needs a minimum modification in implementing on any mobile robot platform that can be used in robotic research, education, laboratories, office environment and factory.