A Dijkstra-Inspired Graph Algorithm for Fully Autonomous Tasking in Industrial Applications
Mohamed Lotfi,Gerardo J. Osrio,Mohammad S. Javadi,Abdelrahman Ashraf,Mustafa Zahran,Georges Samih,Joo P. S. Catalo,Gerardo Osorio,Mohammad Javadi,Joao P. S. Catalao
DOI: https://doi.org/10.1109/tia.2021.3091418
IF: 4.079
2021-09-01
IEEE Transactions on Industry Applications
Abstract:An original graph-based model and algorithm for optimal industrial task scheduling is proposed in this article. The innovative algorithm designed, dubbed "Dijkstra optimal tasking" (DOT), is suitable for fully distributed task scheduling of autonomous industrial agents for optimal resource allocation, including energy use. The algorithm was designed starting from graph theory fundamentals, from the ground up, to guarantee a generic nature, making it applicable on a plethora of tasking problems and not case-specific. For any industrial setting in which mobile agents are responsible for accomplishing tasks across a site, the objective is to determine the optimal task schedule for each agent, which maximizes the speed of task achievement while minimizing the movement, thereby minimizing energy consumption cost. The DOT algorithm is presented in detail in this manuscript, starting from the conceptualization to the mathematical formulation based on graph theory, having a thorough computational implementation and a detailed algorithm benchmarking analysis. The choice of Dijkstra as opposed to other shortest path methods (namely, A* Search and Bellman–Ford) in the proposed graph-based model and algorithm was investigated and justified. An example of a real-world application based on a refinery site is modeled and simulated and the proposed algorithm's effectiveness and computational efficiency is duly evaluated. A dynamic obstacle course was incorporated to effectively demonstrate the proposed algorithm's applicability to real-world applications.
engineering, electrical & electronic, multidisciplinary