Razvoj sistema za vodenje flote industrijskih samodejno vodenih vozičkov

Redenšek, Kristina (author), Vrabič, Rok (mentor)

V sklopu industrije 4.0, ki je poimenovanje za trenuten trend avtomatizacije in izmenjave podatkov v proizvodnih sistemih, velik razvoj doživlja tudi področje samodejno vodenih vozičkov (AGV). Znotraj tega magistrskega dela smo zasnovali programsko opremo za krmiljenje sistema takšnih vozil. Krmiljenje obsega dodeljevanje nalog, načrtovanje poti in časovno usklajevanje poti več vozil. Implementacija programske opreme sloni na uporabi vozliščnih grafov in metode s časovnimi okni. Programska oprema je bila testirana v simulacijskem okolju za več različnih velikosti delovnega okolja sistema AGV-jev in več različnih števil nalog. Rezultati kažejo, da je programska oprema zadovoljivo hitra in uspešna pri načrtovanju krmiljenja do 100 nalog, saj lahko te načrtuje v 5 sekundah s 100% uspešnostjo. Izdelana programska oprema je matematično razumljiva, fleksibilna, skalabilna in modularna in glede na rezultate primerna za praktično uporabo. Razvit algoritem je bil tudi implementiran v obstoječe ROS (ang. Robot Operating System) ogrodje.

• Language: Slovenian
• Keywords: AGV, intralogistika, upravljanje flote, vozliščni grafi, časovna okna, ROS
• Work type: Master's thesis/paper
• Typology: 2.09 - Master's Thesis
• Organization: FS - Faculty of Mechanical Engineering
• Place of publishing: Ljubljana
• Publisher: [K. Redenšek]
• Year: 2019
• Number of pages: XXI, 62 str.
• PID: 20.500.12556/RUL-109379
• UDC: 004.451.25(043.2)
• COBISS.SI-ID: 16864795
• Publication date in RUL: 31.08.2019

Secondary language

• Language: English
• Title: Development of a fleet management system for industrial automated guided vehicles

Abstract

As a part of Industry 4.0, which is the name for the current trend of automation and data exchange in manufacturing systems, there is a significant development in the field of autonomous guided vehicles (AGV). Within this master thesis, we developed software for control of AGV systems. The control covers dispatching, routing and scheduling of vehicles' tasks. An implementation of the software is based on node graphs and the time-window method. The software was tested in a simulation environment for different scales of the working environment, and a different number of tasks. The results show that the software is sufficiently fast. It can plan up to 100 tasks within 5 seconds with 100% success rate. The result of our work is a mathematically sound, flexible, scalable, and modular solution. According to the results, the solution can be used in real-world cases. The algorithm was also implemented using the Robot Operating System (ROS) framework.

• Keywords: AGV, intralogistics, fleet management, node graphs, time windows, ROS