AGV precision docking in smart factory

• Main Author: Tan, Jia Yi
• Other Authors: Xie Lihua
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2021
• Subjects: Engineering::Electrical and electronic engineering
• Online Access: https://hdl.handle.net/10356/149223
• Institution: Nanyang Technological University

Automated guided vehicle (AGV) has become a research hotspot in recent years because of its great performance in executing repetitive tasks. The main concerns of the current methodology are to maintain the high precision of AGV docking after being maneuvered to multiple goals and to minimize human intervention in the process of docking. Therefore, this project proposes the implementation of Ultra Wideband (UWB) coordinating system to supplement the vision-based docking mechanism using LiDAR sensor alone. In recent studies, UWB sensor is becoming more popular because it is able to provide precise localization and cover a larger operating area as compared to other wireless positioning technologies. Therefore, a UWB based navigation algorithm is developed in this project using the concept of Potential Field Method to perform tag following, collision avoidance and path tracking tasks. This algorithm can navigate the AGV to the goal with maximum linear error of 40 centimeters and maximum angular error of 12 degrees. Other than the navigation algorithm, UWB sensors are also used to perform automated map scanning and provide the initial pose estimate for the adaptive Monte Carlo localization (AMCL) so that the AGV can be successfully initialized at any position in the map. After the localization process, a ROS package “move_base” is used to perform AGV precision docking to single station and multiple stations. Eventually, the AGV can be docked to both the dynamic station and static station with comparable accuracy around 25 centimeters and to multiple linearly oriented stations with deviation lower than 15 centimeters.