Language Support for Multi Agent Reinforcement Learning

Abstract

Full bibliographic details must be given when referring to, or quoting from full items including the author's name, the title of the work, publication details where relevant (place, publisher, date), pagination, and for theses or dissertations the awarding institution, the degree type awarded, and the date of the award.

Digital Twins: Review and Challenges

“…Regarding the level of integration, Kritzinger in his work [50], get as most used level the Digital Shadow (35 percent), followed by Digital Model (28 percent) and with a lower percentage the Digital Twin (18 percent), having a 19 percent of publications of indeterminate level of integration. The publications that present data exchange do so through the main communication trends, addressed in a previous section, however, only the publications [71,15,86,92,20,5,66,52,44,4,51,35,75,23] present communication between DTs of the same system. The results of the categorization can be seen in Fig.…”

“…The authors use international standards for their platform, called uDiT (universal Digital Twin platform), and presents the next elements: a middleware supported by OMG (Object Management Group) and DDS (Data Distribution Service) tools; besides allows co-simulation through FMI, and presents gateway methods. Others examples that belong to the informatics area and that have been previously described can be seen in [81,3,66,44,52,75,23].…”

“…Clark [23] creates a language expansion which includes agents with learning and operations capabilities, the authors also present its performance (ESL platform developed by the authors), describe its mains characteristics, and present a case study base on a supply chain system.…”

“…Although all articles use the term Digital Twin within their content, not all of them present the bi-directional exchange of data among the DT and the physical twin. The publications that present data exchange do so through the main communication trends, addressed in a previous section, however, only the publications[71,15,86,92,20,5,66,52,44,4,51,35,75,23] present communication between DTs of the same system. The results of the categorization can be seen in Fig.24.…”

Digital twin-driven smart supply chain

“…Considering that supply chains are becoming increasingly complex, several researchers discuss how to deal with the complexity and enable intelligent supply chain in the DT context. Simulation tools are useful to represent the interdependencies in complex supply chains in the DT context, including discrete event simulation (D'Angelo and Chong, 2018; Dobler et al, 2020;Dutta et al, 2021;Pilati et al, 2021;Wilson et al, 2021), agent-based simulation (Gorodetsky et al, 2019;Clark et al, 2020;Orozco-Romero et al, 2020), and hybrid simulation models (Makarov et al, 2021). Frazzon et al (2020) address distributed decision-making under a socio-cyberphysical system perspective in the DTSC context.…”